Aerial work vehicle metal parts quality analysis system based on machine vision

Abstract

The present invention belongs to the technical field of metal fittings quality analysis, and discloses a quality analysis system for metal fittings of aerial work vehicles based on machine vision, the system comprising an image acquisition module, a quality anomaly analysis module, a quality change trend analysis module, a quality hidden danger judgment module and a data storage library. The present invention analyzes the structural stability, dimensional consistency and apparent integrity of the target metal fittings based on the acquired images of the target metal fittings, thereby analyzing the quality anomalies of the target metal fittings. This analysis method improves the comprehensiveness of the quality analysis results and the accuracy of risk assessment, and ensures the effect of quality analysis. The present invention conducts comprehensive safety hazard identification through the current quality situation and the changing quality situation based on historical monitoring records. This analysis method improves the continuity and predictability of quality analysis and risk assessment, avoids unpredictable safety accidents, and ensures construction progress and personnel safety.

Description

Technical Field

[0001] The invention belongs to the technical field of metal fittings quality analysis, and relates to a metal fittings quality analysis system for an aerial work vehicle based on machine vision. Background Art

[0002] Aerial work vehicles are a type of engineering machinery used for aerial work, mainly for construction, maintenance, and overhaul. The metal parts of aerial work vehicles usually include outriggers, booms, work platforms, and connectors. The work platform is an important component of the aerial work vehicle, and it undertakes the important task of ensuring the smooth implementation of the project and the safety of construction personnel. The quality analysis of traditional aerial work vehicle metal parts is usually lagging and one-sided, so the quality analysis of aerial work vehicle metal parts based on machine vision is of great significance.

[0003] Traditional quality analysis of aerial work vehicle metal accessories usually only conducts qualitative analysis on a single monitoring indicator, ignoring the overall quality and comprehensive performance evaluation of aerial work vehicle metal accessories. This analysis method causes one-sidedness in quality analysis results and inaccuracy in risk assessment, and cannot fully reflect the true overall quality status of metal accessories. At the same time, when a fault occurs, it is difficult to quickly locate the problem based on a single quality monitoring indicator.

[0004] In the traditional quality analysis process of aerial work vehicle metal accessories, the analysis of metal accessories quality is usually based on the analysis and judgment of existing or occurred problems, while ignoring the risk assessment of the changing trends of various monitoring indicators. This analysis method reduces the continuity and predictability of quality analysis and risk assessment. It only focuses on the quality of the current metal accessories and cannot accurately analyze the actual quality changes of metal accessories. It may cause unpredictable safety accidents, affect the construction progress, and also bring hidden dangers to personnel safety. Summary of the invention

[0005] In view of this, in order to solve the problems of lag and one-sidedness in the quality analysis of metal parts of traditional aerial work vehicles mentioned in the above background technology, a quality analysis system of metal parts of aerial work vehicles based on machine vision is proposed.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A machine vision-based aerial work vehicle metal accessories quality analysis system includes: an image acquisition module, which is used to record the metal structure accessories of the current aerial work vehicle's working platform as target metal accessories, and then obtain the captured image of the target metal accessories.

[0007] The quality anomaly analysis module is used to analyze the structural stability, dimensional consistency and apparent integrity of the target metal accessories based on the collected images of the target metal accessories, thereby analyzing the quality anomalies of the target metal accessories.

[0008] The quality change trend analysis module is used to analyze the quality change trend data of the target metal accessories based on the historical quality monitoring records, and then analyze the abnormal situation of the quality change trend of the target metal accessories.

[0009] The quality hazard judgment module is used to judge whether the target metal accessories have safety hazards based on the quality abnormalities and quality change trend abnormalities of the target metal accessories. If so, the specific hazard type is further identified. The hazard type includes current hazards and quality trend hazards, and different processing operations are performed for different hazard types.

[0010] A data repository for storing historical quality monitoring records.

[0011] In a preferred embodiment of the present invention, the stability of the architecture is specifically analyzed as follows: extracting a number of captured images of the target metal accessories captured by a high-definition camera, and obtaining the capture time of each captured image.

[0012] Image processing technology is used to obtain the support point position of the bottom of the target metal accessories in each collected image, the center of gravity position of each load, and the height and bottom area corresponding to each load, and the tipping distance deviation index of the target metal accessories is obtained based on this analysis. .

[0013] The acquisition time of each acquired image is compared with the acquisition time of the previous acquired image to obtain the acquisition interval duration of each acquired image, and the image processing software is used to obtain the bottom inclination angle of the target metal accessory of each acquired image. Then, the bottom inclination angle of the target metal accessory of each acquired image is compared with the bottom inclination angle of the target metal accessory of the previous acquired image to obtain the change in the bottom inclination angle of the target metal accessory of each acquired image.

[0014] The vibration amplitude per unit time of each collected image is obtained by calculating the ratio of the bottom tilt angle change of the target metal accessory of each collected image to the corresponding collection interval of each collected image. The vibration amplitude per unit time of each collected image is compared with the preset vibration amplitude threshold per unit time to obtain the vibration index of the target metal accessory of each collected image, and then the average is calculated to obtain the vibration index of the target metal accessory. .

[0015] The tipping distance deviation index of the target metal fittings and vibration index Calculate the structural stability index of the target metal fittings according to the weight .

[0016] In a preferred embodiment of the present invention, the specific analysis method of the tipping distance deviation index is as follows: extracting the support point position of the bottom of the target metal accessory, the center of gravity position of each load, and the height and bottom area corresponding to each load in each collected image obtained by image processing technology.

[0017] Connect the positions of adjacent support points to obtain several tipping lines, each tipping line constitutes a safe area, connect the center of gravity position of each load with the center of gravity of the target metal accessory and obtain the distance of the corresponding line segment as the center of gravity deviation distance of each load.

[0018] The center of gravity of each load is connected vertically to each tipping line, the length of each vertical line segment is obtained and compared, and the length of the vertical line segment with the largest length is selected and recorded as the tipping distance.

[0019] The tipping distance of each load on the target metal fitting in each acquired image ,high and bottom area And the center of gravity deviation distance of each load Substitute into the formula Get the tipping distance deviation index of the target metal parts in each collected image ,in Indicates the load number, , Indicates the maximum guardrail height of the target metal fitting, Indicates the bottom area of ​​the target metal fitting, Indicates the bottom length of the target metal fitting. Respectively represent the influence weights of the tipping distance and the center of gravity deviation distance, .

[0020] The tipping distance deviation index of the target metal parts in each collected image The mean value is calculated to obtain the tipping distance deviation index of the target metal fittings. .

[0021] In a preferred embodiment of the present invention, the size consistency is specifically analyzed as follows: extract a number of captured images of the target metal accessories captured by a high-definition camera, use image processing technology to obtain the gear lever contours of the target metal accessories in each captured image to form gear lever contour data, and then compare with the reference target metal accessories gear lever contour data.

[0022] The included angles of each gear lever and the corresponding cross gear lever are compared with the corresponding standard gear lever included angle in the gear lever profile data of the reference target metal accessory to obtain the gear lever included angle deviation corresponding to each gear lever, and then the mean value is calculated to obtain the gear lever included angle deviation of the target metal accessory in each collected image. .

[0023] Obtain the length and width of each lever in each collected image, classify each lever according to its width to obtain the lever corresponding to each width, and then calculate the sum of the lengths to obtain the actual lever length corresponding to the lever of each width At the same time, based on the reference target metal accessory gear lever profile data, the standard gear lever length corresponding to each width gear lever is obtained. .

[0024] The actual gear rod length corresponding to each width gear rod in each acquired image , Standard gear lever length Deviation between the gear lever angle and the Substitute into the analysis formula Get the size consistency index of the target metal accessories in each collected image ,in Indicates the width number, , Indicates The width value corresponding to the class width, Indicates the maximum lever width, , and then calculate the mean to obtain the dimensional consistency index of the target metal accessories .

[0025] In a preferred embodiment of the present invention, the apparent integrity is specifically analyzed as follows: extract a number of acquired images of the target metal accessories acquired by a high-definition camera, and use image processing technology to obtain the defect area of ​​the target metal accessories and the surface area of ​​the target metal accessories in each acquired image.

[0026] The defective area of ​​the target metal accessory in each acquired image is compared with the corresponding surface area of ​​the target metal accessory to obtain the apparent integrity index of the target metal accessory in each acquired image.

[0027] Compare the apparent integrity index of the target metal accessories in each collected image, and select the smallest apparent integrity index as the apparent integrity index of the target metal accessories .

[0028] In a preferred embodiment of the present invention, the analysis of the quality abnormality of the target metal fitting is specifically as follows: the structural stability index of the target metal fitting is , Dimensional consistency index and apparent integrity index The quality anomaly index of the target metal accessories is obtained by calculation according to the weight.

[0029] In a preferred embodiment of the present invention, the quality change trend data is specifically analyzed as follows: historical quality monitoring records are extracted, and the monitoring index quantities corresponding to each historical quality monitoring record and the monitoring time corresponding to each historical quality monitoring record are obtained, and the monitoring index quantities include an architectural stability index, a dimensional consistency index and an apparent integrity index.

[0030] A number of monitoring indicator change curves are drawn with the monitoring time as the horizontal axis and each monitoring indicator as the vertical axis. A number of monitoring points are evenly arranged on each monitoring indicator change curve to obtain the curve slope of each monitoring point in each monitoring indicator change curve.

[0031] In a preferred embodiment of the present invention, the analysis of abnormal quality change trends of target metal accessories is specifically as follows: the slope of the curve of each monitoring point in the change curve of each monitoring indicator is extracted and compared to obtain the change rate of each monitoring indicator, and then the difference with the pre-set change rate threshold is calculated to obtain the change rate deviation of each monitoring indicator, and the change rate deviation corresponding to each monitoring indicator is calculated by ratio with the change rate threshold to obtain the quality change trend abnormality index of each monitoring indicator.

[0032] The quality change trend abnormality indexes corresponding to the monitoring indicators of the target metal accessories are compared, and the largest quality change trend abnormality index is selected as the quality change trend abnormality index of the target metal accessories.

[0033] In a preferred embodiment of the present invention, the determination of whether a target metal accessory has a safety hazard is specifically as follows: extracting a quality anomaly index and a quality change trend anomaly index of the target metal accessory, and then comparing them with a pre-set quality anomaly index threshold and a quality change trend anomaly index threshold, respectively, to obtain a determination result of whether a target metal accessory has a safety hazard.

[0034] In a preferred embodiment of the present invention, the identification of specific hidden danger types is as follows: when it is determined that there is a safety hazard, the target metal accessory is recorded as a hidden danger metal accessory, and then the quality anomaly index of the hidden danger metal accessory is compared with the quality anomaly index threshold.

[0035] If the quality anomaly index of the metal accessories with hidden dangers is less than or equal to the quality anomaly index threshold, the specific hidden danger type is judged to be a quality trend hidden danger; if the quality anomaly index of the metal accessories with hidden dangers is greater than the quality anomaly index threshold, the specific hidden danger type is judged to be a current hidden danger.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) When performing the current quality abnormality analysis of the target metal accessories, the present invention analyzes the structural stability, dimensional consistency and apparent integrity of the target metal accessories based on the captured images of the target metal accessories, thereby analyzing the quality abnormality of the target metal accessories. This analysis method improves the comprehensiveness of the quality analysis results and the accuracy of risk assessment, and can fully reflect the true overall quality status of the metal accessories. At the same time, when a fault occurs, multiple quality monitoring indicator analyses can quickly locate the problem, thereby ensuring the effect of quality analysis.

[0037] (2) When judging the quality hazards of target metal accessories, the present invention conducts comprehensive safety hazard identification through the current quality status and the quality change status based on historical monitoring records. This analysis method improves the continuity and predictability of quality analysis and risk assessment. The combined analysis of the current quality status and quality change trend of the target metal accessories can accurately obtain the actual quality status of the metal accessories, avoid unpredictable safety accidents, and ensure the construction progress and personnel safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0039] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1 As shown, the present invention provides a quality analysis system for metal accessories of aerial work vehicles based on machine vision, the system includes an image acquisition module, a quality anomaly analysis module, a quality change trend analysis module, a quality hidden danger judgment module and a data storage library, wherein the image acquisition module is connected to the quality anomaly analysis module, the quality anomaly analysis module and the quality change trend analysis module are both connected to the quality hidden danger judgment module, and the data storage library is connected to the quality change trend analysis module.

[0042] The image acquisition module is used to record the metal structural accessories of the current working platform of the aerial work vehicle as target metal accessories, and then obtain the acquisition image of the target metal accessories.

[0043] The quality anomaly analysis module is used to analyze the structural stability, dimensional consistency and apparent integrity of the target metal accessories based on the collected images of the target metal accessories, thereby analyzing the quality anomalies of the target metal accessories.

[0044] It should be noted that the reasons for selecting the structural stability, dimensional consistency and apparent integrity of the target metal accessories as factors affecting quality abnormalities are: 1. Structural stability is directly related to the safety of the aerial work vehicle during use. If the structure of the metal accessories is unstable, it may cause the equipment to become unstable or even collapse when under load. 2. Dimensional consistency affects the fitting accuracy between metal accessories, which in turn affects the function and performance of the entire equipment. 3. Maintaining the apparent integrity helps reduce the risk of corrosion, especially for aerial work vehicles working in harsh environments. Good apparent integrity helps improve the durability and service life of metal accessories.

[0045] Preferably, the stability of the architecture is specifically analyzed as follows: extracting a number of captured images of the target metal accessories captured by a high-definition camera, and obtaining the capture time of each captured image.

[0046] Image processing technology is used to obtain the support point position of the bottom of the target metal accessories in each collected image, the center of gravity position of each load, and the height and bottom area corresponding to each load, and the tipping distance deviation index of the target metal accessories is obtained based on this analysis. .

[0047] For example, the load may be construction materials or construction workers.

[0048] The acquisition time of each acquired image is compared with the acquisition time of the previous acquired image to obtain the acquisition interval duration of each acquired image, and the image processing software is used to obtain the bottom inclination angle of the target metal accessory of each acquired image. Then, the bottom inclination angle of the target metal accessory of each acquired image is compared with the bottom inclination angle of the target metal accessory of the previous acquired image to obtain the change in the bottom inclination angle of the target metal accessory of each acquired image.

[0049] It should be supplemented that the acquisition interval duration of each acquired image is obtained by performing a difference calculation between the acquisition time of each acquired image and the acquisition time of the previous acquired image, and the change in the inclination angle of the bottom of the target metal accessory of each acquired image is obtained by performing a difference calculation between the inclination angle of the bottom of the target metal accessory of each acquired image and the inclination angle of the bottom of the target metal accessory of the previous acquired image and then taking the absolute value.

[0050] It should be further supplemented that the collected images are imported into the image processing software for grayscale, binarization and denoising, and then feature extraction and feature location are performed. Finally, the angle of the corresponding target feature can be obtained by measuring the feature angle.

[0051] It should be noted that the first acquired image does not have the corresponding acquisition interval duration and the change in the bottom tilt angle of the target metal accessory. Therefore, in the subsequent analysis, only the other acquired images except the first acquired image are analyzed.

[0052] The vibration amplitude per unit time of each collected image is obtained by calculating the ratio of the bottom tilt angle change of the target metal accessory of each collected image to the corresponding collection interval of each collected image. The vibration amplitude per unit time of each collected image is compared with the preset vibration amplitude threshold per unit time to obtain the vibration index of the target metal accessory of each collected image, and then the average is calculated to obtain the vibration index of the target metal accessory. .

[0053] It should be supplemented that the vibration amplitude per unit time of each acquired image is subtracted from the preset vibration amplitude per unit time threshold to obtain the vibration amplitude deviation per unit time of each acquired image, and then the vibration index of the target metal accessories of each acquired image is obtained by ratio calculation with the vibration amplitude per unit time threshold.

[0054] The tipping distance deviation index of the target metal fittings and vibration index Calculate the structural stability index of the target metal fittings according to the weight .

[0055] Exemplarily, the weights of the rollover distance deviation index and the vibrancy index are 0.65 and 0.35.

[0056] It should be noted that the deviation of the tipping distance and the vibration condition affect the structural stability of the target metal accessories: 1. The tipping distance refers to the support line of the equipment at the moment when it loses stability and is about to tip over, also known as the tipping line or tipping edge. The deviation of the tipping distance will make the equipment more prone to tipping under certain conditions, especially when the equipment is in an unbalanced state. If the tipping distance deviates, it may cause the center of gravity of the equipment to shift, thereby affecting the overall stability of the equipment. 2. Long-term vibration can cause fatigue damage to metal accessories, especially in stress concentration areas such as welds or connection points. Vibration may cause fasteners (such as bolts and screws) to loosen, thereby affecting the structural integrity and stability of the equipment.

[0057] Preferably, the specific analysis method of the tipping distance deviation index is as follows: extracting the support point position of the bottom of the target metal accessory, the center of gravity position of each load, and the height and bottom area corresponding to each load in each collected image obtained by using image processing technology.

[0058] Connect the positions of adjacent support points to obtain several tipping lines, each tipping line constitutes a safe area, connect the center of gravity position of each load with the center of gravity of the target metal accessory and obtain the distance of the corresponding line segment as the center of gravity deviation distance of each load.

[0059] The center of gravity of each load is connected vertically to each tipping line, the length of each vertical line segment is obtained and compared, and the length of the vertical line segment with the largest length is selected and recorded as the tipping distance.

[0060] It should be noted that a large vertical line segment length can reflect the deviation of the center of gravity. The larger the vertical line segment length, the more serious the deviation of the center of gravity position.

[0061] The tipping distance of each load on the target metal fitting in each acquired image ,high and bottom area And the center of gravity deviation distance of each load Substitute into the formula Get the tipping distance deviation index of the target metal parts in each collected image ,in Indicates the load number, , Indicates the maximum guardrail height of the target metal fitting, Indicates the bottom area of ​​the target metal fitting, Indicates the bottom length of the target metal fitting. Respectively represent the influence weights of the tipping distance and the center of gravity deviation distance, .

[0062] For example, .

[0063] The tipping distance deviation index of the target metal parts in each collected image The mean value is calculated to obtain the tipping distance deviation index of the target metal fittings. .

[0064] Preferably, the size consistency is specifically analyzed as follows: extract a number of captured images of the target metal accessories captured by a high-definition camera, use image processing technology to obtain the gear lever contours of the target metal accessories in each captured image to form gear lever contour data, and then compare with the reference target metal accessories gear lever contour data.

[0065] The included angles of each gear lever and the corresponding cross gear lever are compared with the corresponding standard gear lever included angle in the gear lever profile data of the reference target metal accessory to obtain the gear lever included angle deviation corresponding to each gear lever, and then the mean value is calculated to obtain the gear lever included angle deviation of the target metal accessory in each collected image. .

[0066] It should be supplemented that the gear lever angle deviation corresponding to each gear lever is obtained by performing difference calculation between the angle of each cross gear lever and the corresponding standard gear lever angle in the gear lever profile data of the reference target metal accessory and then taking the absolute value.

[0067] Obtain the length and width of each lever in each collected image, classify each lever according to its width to obtain the lever corresponding to each width, and then calculate the sum of the lengths to obtain the actual lever length corresponding to the lever of each width At the same time, based on the reference target metal accessory gear lever profile data, the standard gear lever length corresponding to each width gear lever is obtained. .

[0068] It should be noted that the reason for classifying and analyzing each lever according to width is that width is an important parameter for measuring the supporting capacity of the lever. A larger width can provide better support and stability. In practice, the larger the lever width, the greater the supporting force it bears, which is often more important for the stability of the target metal accessories.

[0069] The actual gear rod length corresponding to each width gear rod in each acquired image , Standard gear lever length Deviation between the gear lever angle and the Substitute into the analysis formula Get the size consistency index of the target metal accessories in each collected image ,in Indicates the width number, , Indicates The width value corresponding to the class width, Indicates the maximum lever width, , and then calculate the mean to obtain the dimensional consistency index of the target metal accessories .

[0070] Preferably, the apparent integrity is specifically analyzed as follows: extracting several acquired images of the target metal accessory acquired by a high-definition camera, and using image processing technology to obtain the defective region area of ​​the target metal accessory and the surface area of ​​the target metal accessory in each acquired image.

[0071] Exemplarily, the defects may be rust, dents, and surface shedding, etc.

[0072] The defective area of ​​the target metal accessory in each acquired image is compared with the corresponding surface area of ​​the target metal accessory to obtain the apparent integrity index of the target metal accessory in each acquired image.

[0073] It should be supplemented that the defect area of ​​the target metal accessories in each acquired image is calculated by ratioing the defect area of ​​the target metal accessories in each acquired image to the corresponding surface area of ​​the target metal accessories, and then the defect area ratio of the target metal accessories in each acquired image is substituted into the formula Get the apparent integrity index of the target metal parts in each collected image ,in Indicates the number of the acquired image. .

[0074] Compare the apparent integrity index of the target metal accessories in each collected image, and select the smallest apparent integrity index as the apparent integrity index of the target metal accessories .

[0075] It should be noted that the minimum apparent integrity index is selected as the apparent integrity index of the target metal accessory because in practice, the collected image may not be able to fully cover the entire surface area of ​​the target metal accessory. Therefore, selecting the minimum apparent integrity index as the apparent integrity index of the target metal accessory can more truly reflect the apparent integrity of the target metal accessory.

[0076] Preferably, the analysis of the quality abnormality of the target metal fitting is as follows: the structural stability index of the target metal fitting is , Dimensional consistency index and apparent integrity index The quality anomaly index of the target metal accessories is obtained by calculation according to the weight.

[0077] It should be added that the quality abnormality index of the target metal accessories is specifically analyzed as follows: the structural stability index of the target metal accessories , Dimensional consistency index and apparent integrity index Substitute into the formula Get the quality anomaly index of the target metal parts ,in Represent the weights of the architectural stability index, dimensional consistency index and apparent integrity index respectively.

[0078] For example, .

[0079] It needs to be explained that the reasons for choosing the structural stability index, dimensional consistency index and apparent integrity index as influencing factors of the quality anomaly index of the target metal accessories are: 1. The structural stability index reflects the structural stability and ability to resist deformation of metal accessories when subjected to external loads. Structural stability is crucial to ensuring the structural integrity and safety of metal accessories under normal use conditions. If the structure of the metal accessories is unstable, it may cause unexpected deformation or damage during use, affecting its functional performance; 2. The dimensional consistency index measures the consistency of the dimensions of each part of the metal accessories during the manufacturing process. Dimensional consistency is critical to ensuring that the metal accessories meet the expected functional and performance standards; 3. The apparent integrity index focuses on the integrity of the appearance of metal accessories, including surface finish, coating quality and other aspects. Metal accessories with good surface treatment are more resistant to corrosion and wear and extend their service life.

[0080] It should be noted that the present invention analyzes the quality abnormality of the target metal accessories by analyzing the structural stability, dimensional consistency and apparent integrity of the target metal accessories based on the captured images of the target metal accessories when conducting the current quality abnormality analysis. This analysis method improves the comprehensiveness of the quality analysis results and the accuracy of risk assessment, and can fully reflect the true overall quality status of the metal accessories. At the same time, when a fault occurs, analysis of multiple quality monitoring indicators can quickly locate the problem, thereby ensuring the effect of quality analysis.

[0081] The quality change trend analysis module is used to analyze the quality change trend data of the target metal accessories based on the historical quality monitoring records, and then analyze the abnormal situation of the quality change trend of the target metal accessories.

[0082] It should be noted that the quality change trend refers to the quality change trend of the target metal accessories obtained from the historical quality monitoring records, which reflects the quality change tendency of the target metal accessories. In practice, the quality change trend of the target metal accessories should be within the controllable range. When the quality change trend of the target metal accessories exceeds the controllable range, it indicates that there are major safety hazards in the quality of the target metal accessories.

[0083] Preferably, the quality change trend data is specifically analyzed as follows: extract historical quality monitoring records, and obtain the monitoring indicator quantities corresponding to each historical quality monitoring record and the monitoring time corresponding to each historical quality monitoring record, wherein the monitoring indicator quantities include an architectural stability index, a dimensional consistency index and an apparent integrity index.

[0084] A number of monitoring indicator change curves are drawn with the monitoring time as the horizontal axis and each monitoring indicator as the vertical axis. A number of monitoring points are evenly arranged on each monitoring indicator change curve to obtain the curve slope of each monitoring point in each monitoring indicator change curve.

[0085] It needs to be explained that the slope of the curve at each monitoring point reflects the specific situation of the quality change trend of the target metal accessories at that monitoring point.

[0086] Preferably, the analysis of abnormal quality change trends of target metal accessories is as follows: the slope of the curve of each monitoring point in the change curve of each monitoring indicator is extracted and compared to obtain the change rate of each monitoring indicator, and then the difference with the pre-set change rate threshold is calculated to obtain the change rate deviation of each monitoring indicator, and the change rate deviation corresponding to each monitoring indicator is calculated by ratio with the change rate threshold to obtain the quality change trend abnormality index of each monitoring indicator.

[0087] Exemplarily, the rate of change threshold is -0.02.

[0088] It should be added that the change rate of each monitoring indicator quantity of the target metal fittings is obtained by calculating the average of the curve slopes of each monitoring point in the change curve of each monitoring indicator quantity.

[0089] The quality change trend abnormality indexes corresponding to the monitoring indicators of the target metal accessories are compared, and the largest quality change trend abnormality index is selected as the quality change trend abnormality index of the target metal accessories.

[0090] It should be noted that the reason for selecting the largest quality change trend abnormality index as the quality change trend abnormality index of the target metal accessories is: if the largest quality change trend abnormality index is also within the controllable range, it means that the quality change trends of each monitoring indicator of the target metal accessories are controllable.

[0091] The quality hazard judgment module is used to judge whether the target metal accessories have safety hazards based on the quality abnormalities and quality change trend abnormalities of the target metal accessories. If so, the specific hazard type is further identified. The hazard type includes current hazards and quality trend hazards, and different processing operations are performed for different hazard types.

[0092] Preferably, the determination of whether a target metal accessory has a safety hazard is specifically as follows: extracting a quality anomaly index and a quality change trend anomaly index of the target metal accessory, and then comparing them with a pre-set quality anomaly index threshold and a quality change trend anomaly index threshold, respectively, to obtain a determination result of whether a target metal accessory has a safety hazard.

[0093] It should be added that the process of judging whether the target metal accessories have safety hazards is as follows: if the quality abnormality index of the target metal accessories is greater than the quality abnormality index threshold or the quality change trend abnormality index is greater than the quality change trend abnormality index threshold, then it is judged that the target metal accessories have safety hazards; if the quality abnormality index of the target metal accessories is less than or equal to the quality abnormality index threshold and the quality change trend abnormality index is less than or equal to the quality change trend abnormality index threshold, then it is judged that the target metal accessories do not have safety hazards.

[0094] Exemplarily, the quality anomaly index threshold and the quality change trend anomaly index threshold are 0.15 and 0.3 respectively.

[0095] Preferably, the identification of a specific hidden danger type is specifically as follows: when it is determined that there is a safety hazard, the target metal accessory is recorded as a hidden danger metal accessory, and then the quality anomaly index of the hidden danger metal accessory is compared with the quality anomaly index threshold.

[0096] It should be noted that in practice, the current quality of the target metal accessories is not abnormal, but there is an abnormal tendency, so the specific situation needs to be analyzed.

[0097] If the quality anomaly index of the metal accessories with hidden dangers is less than or equal to the quality anomaly index threshold, the specific hidden danger type is judged to be a quality trend hidden danger; if the quality anomaly index of the metal accessories with hidden dangers is greater than the quality anomaly index threshold, the specific hidden danger type is judged to be a current hidden danger.

[0098] It should be noted that different processing operations are performed for different hidden danger types.

[0099] Exemplarily, when the specific hidden danger type is a quality trend hidden danger, the image acquisition frequency is increased, and when the specific hidden danger type is a current hidden danger, the construction operation is stopped.

[0100] It should be noted that the present invention conducts comprehensive safety hazard identification through the current quality status and the changing quality status based on historical monitoring records when judging the quality hazards of target metal accessories. This analysis method improves the continuity and predictability of quality analysis and risk assessment. The combined analysis of the current quality status and quality change trend of the target metal accessories can accurately obtain the actual quality status of the metal accessories, avoid unpredictable safety accidents, and ensure construction progress and personnel safety.

[0101] The data repository is used to store historical quality monitoring records.

[0102] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. Aerial work vehicle metal parts quality analysis system based on machine vision, characterized by: include: An image acquisition module is used to record the metal structural parts of the current working platform of the aerial work vehicle as target metal parts, and then obtain the acquisition image of the target metal parts; A quality anomaly analysis module is used to analyze the structural stability, dimensional consistency and apparent integrity of the target metal accessories based on the collected images of the target metal accessories, thereby analyzing the quality anomalies of the target metal accessories; Extracting a number of captured images of the target metal accessories captured by a high-definition camera, and obtaining the acquisition time of each captured image; The tipping distance deviation index of the target metal fittings and vibration index Calculate the structural stability index of the target metal fittings according to the weight ; Dimensional consistency index of target metal parts in each acquired image The actual gear rod length corresponding to each width gear rod in each collected image , Standard gear lever length Deviation between the gear lever angle and the Substitute into the analysis formula Get, where Indicates the width number, , Indicates The width value corresponding to the class width, Indicates the maximum lever width, , and then calculate the mean to obtain the dimensional consistency index of the target metal accessories ; The specific abnormalities in the quality of the target metal parts are as follows: The structural stability index of the target metal fittings , Dimensional consistency index and apparent integrity index The quality anomaly index of the target metal fittings is obtained by calculating according to the weight; The quality change trend analysis module is used to analyze the quality change trend data of the target metal accessories based on the historical quality monitoring records, and then analyze the abnormality of the quality change trend of the target metal accessories; A quality hidden danger judgment module is used to judge whether the target metal accessories have safety hidden dangers based on the abnormal quality and abnormal quality change trend of the target metal accessories. If yes, further identify the specific hidden danger type, which includes current hidden dangers and quality trend hidden dangers, and perform different processing operations for different hidden danger types; A data repository for storing historical quality monitoring records.

2. The machine vision-based aerial work vehicle metal parts quality analysis system according to claim 1, characterized in that: The stability of the architecture is analyzed as follows: Image processing technology is used to obtain the support point position of the bottom of the target metal accessories in each collected image, the center of gravity position of each load, and the height and bottom area corresponding to each load, and the tipping distance deviation index of the target metal accessories is obtained based on this analysis. ; The acquisition time of each acquired image is compared with the acquisition time of the previous acquired image to obtain the acquisition interval of each acquired image, and the bottom tilt angle of the target metal accessory of each acquired image is obtained by using image processing software, and then the bottom tilt angle of the target metal accessory of each acquired image is compared with the bottom tilt angle of the target metal accessory of the previous acquired image to obtain the change amount of the bottom tilt angle of the target metal accessory of each acquired image; The vibration amplitude per unit time of each collected image is obtained by calculating the ratio of the bottom tilt angle change of the target metal accessory of each collected image to the corresponding collection interval of each collected image. The vibration amplitude per unit time of each collected image is compared with the preset vibration amplitude threshold per unit time to obtain the vibration index of the target metal accessory of each collected image, and then the average is calculated to obtain the vibration index of the target metal accessory. .

3. The machine vision-based aerial work vehicle metal parts quality analysis system as claimed in claim 2, characterized in that: The specific analysis method of the tipping distance deviation index is as follows: Extracting the support point position of the bottom of the target metal fitting, the center of gravity position of each load, and the height and bottom area corresponding to each load in each collected image obtained by using image processing technology; Connecting the positions of adjacent support points to obtain several tipping lines, each tipping line constitutes a safe area, connecting the center of gravity position of each load with the center of gravity of the target metal accessory and obtaining the distance of the corresponding line segment as the center of gravity deviation distance of each load; Connect the center of gravity of each load vertically to each tipping line, obtain the length of each vertical line segment and compare them, and select the vertical line segment with the largest length as the tipping distance; The tipping distance of each load on the target metal fitting in each acquired image ,high and bottom area And the center of gravity deviation distance of each load Substitute into the formula Get the tipping distance deviation index of the target metal parts in each collected image ,in Indicates the load number, , Indicates the maximum guardrail height of the target metal fitting, Indicates the bottom area of ​​the target metal fitting, Indicates the bottom length of the target metal fitting. Respectively represent the influence weights of the tipping distance and the center of gravity deviation distance, ; The tipping distance deviation index of the target metal parts in each collected image The mean value is calculated to obtain the tipping distance deviation index of the target metal fittings. .

4. The machine vision-based aerial work vehicle metal parts quality analysis system according to claim 2, characterized in that: The size consistency is analyzed as follows: Extracting a number of acquired images of the target metal accessory acquired by a high-definition camera, using image processing technology to obtain the profile of each gear lever of the target metal accessory in each acquired image to form gear lever profile data, and then comparing it with the reference target metal accessory gear lever profile data; The included angles of each gear lever and the corresponding cross gear lever are compared with the corresponding standard gear lever included angle in the gear lever profile data of the reference target metal accessory to obtain the gear lever included angle deviation corresponding to each gear lever, and then the mean value is calculated to obtain the gear lever included angle deviation of the target metal accessory in each collected image. ; Obtain the length and width of each lever in each collected image, classify each lever according to its width to obtain the lever corresponding to each width, and then calculate the sum of the lengths to obtain the actual lever length corresponding to the lever of each width At the same time, based on the reference target metal accessory gear lever profile data, the standard gear lever length corresponding to each width gear lever is obtained. .

5. The machine vision-based aerial work vehicle metal parts quality analysis system according to claim 4, characterized in that: The detailed analysis of the apparent completeness is as follows: Extracting a number of acquired images of the target metal accessories acquired by a high-definition camera, and using image processing technology to obtain the defect area of ​​the target metal accessories and the surface area of ​​the target metal accessories in each acquired image; Comparing the defect area of ​​the target metal accessory in each collected image with the corresponding surface area of ​​the target metal accessory to obtain the apparent integrity index of the target metal accessory in each collected image; Compare the apparent integrity index of the target metal accessories in each collected image, and select the smallest apparent integrity index as the apparent integrity index of the target metal accessories .

6. The aerial work vehicle metal parts quality analysis system based on machine vision as claimed in claim 1, characterized in that: The quality change trend data is specifically analyzed as follows: Extracting historical quality monitoring records, and obtaining monitoring indicator quantities corresponding to each historical quality monitoring record and monitoring time corresponding to each historical quality monitoring record, wherein the monitoring indicator quantities include an architecture stability index, a dimensional consistency index, and an apparent integrity index; A number of monitoring indicator change curves are drawn with the monitoring time as the horizontal axis and each monitoring indicator as the vertical axis. A number of monitoring points are evenly arranged on each monitoring indicator change curve to obtain the curve slope of each monitoring point in each monitoring indicator change curve.

7. The machine vision-based aerial work vehicle metal parts quality analysis system according to claim 6, characterized in that: The abnormal quality change trend of the target metal parts is as follows: The slope of each monitoring point in the change curve of each monitoring indicator is extracted and compared to obtain the change rate of each monitoring indicator, and then the difference is calculated with the preset change rate threshold to obtain the change rate deviation of each monitoring indicator, and the change rate deviation corresponding to each monitoring indicator is calculated by ratio with the change rate threshold to obtain the quality change trend abnormality index of each monitoring indicator; The quality change trend abnormality indexes corresponding to the monitoring indicators of the target metal accessories are compared, and the largest quality change trend abnormality index is selected as the quality change trend abnormality index of the target metal accessories.

8. The machine vision-based aerial work vehicle metal parts quality analysis system according to claim 7, characterized in that: The specific details of determining whether the target metal fittings have potential safety hazards are as follows: The quality anomaly index and quality change trend anomaly index of the target metal accessories are extracted, and then compared with the pre-set quality anomaly index threshold and quality change trend anomaly index threshold respectively to obtain the judgment result whether the target metal accessories have safety hazards.

9. The machine vision-based aerial work vehicle metal parts quality analysis system according to claim 8, characterized in that: The specific types of hidden dangers identified are as follows: When it is determined that there is a safety hazard, the target metal accessory is recorded as a metal accessory with a hidden danger, and then the quality abnormality index of the metal accessory with a hidden danger is compared with the quality abnormality index threshold; If the quality anomaly index of the metal accessories with hidden dangers is less than or equal to the quality anomaly index threshold, the specific hidden danger type is judged to be a quality trend hidden danger; if the quality anomaly index of the metal accessories with hidden dangers is greater than the quality anomaly index threshold, the specific hidden danger type is judged to be a current hidden danger.

Images